RICHARD E.REEVES
1616
tionating column equipped with a total-condensation variable take-off stillhead. The ethyl alcohol formed in the alcoholysis was fractionated from the mixture at frequent intervals. About 1.5 cc. of material boiling below 150' was separated during the course of the reaction. The excess p-diethylaminoethanol was removed from the reaction mixture by distillation on the steam-bath under reduced pressure, and the residue was dissolved in a small amount of ice-water. The solution was extracted three times with ether and the extracts washed three times with ice-water and dried. On passing hydrogen chloride into the ether solution the product separated as a dark tar which was recrystallized from n-butyl alcohol; m. p. 208O, (dec.); yield, 1.3 g. (17%). Anal. Calcd. for C I S H ~ S O ~ N ~ SC,C 43.3; ~ ~ : H,7.7; N, 8.4; S, 6.4; C1, 21.3. Found: C, 43.9; H, 7.8; N, 8.5; S,6.2; C1, 20.3. 7-Diethylaminopropyl Ester of Thiamorpholine-3,5-dicarboxylic Acid (Trihydrochloride).-This substance was prepared in the same general fashion as just described for the p-diethylaminoethyl ester; m. p. 215" (dec.). Anal. Calcd. for C20H4204NsSC13: N, 7.97. Found: N, 7.98,7.71. Thiamorpholine 3,5 di ( p diethylaminoethyl) carboxamide.-A mixture of 1 g. of the diethyl ester of thiamorpholine-3,5-dicarboxylicacid and 1.5 g. of p-diethylaminoethylamine was heated to 160-170O for twelve hours
-
- - -
[CONTRIBUTION FROM
-
THE
HOSPITAL OF
THE
Vol. 62
in a n open test-tube supported in a Wood's metal bath. Excess amine was removed from the reaction mixture by distillation on the steam-bath under reduced pressure. The residue was dissolved in 50 cc. of dry ether and the product precipitated as the hydrochloride by adding an ethereal solution of hydrogen chloride. The material so obtained was a tan-colored, very hygroscopic, crystalline solid. It was purified by dissolving in hot absolute alcohol, adding ether to the point of turbidity, and allowing to cool. After four such recrystallizations a pure white, practically non-hygroscopic product was obtained which decomposed in the neighborhood of 245O, depending on the rate of heating; yield, 1.1 g. (55%). Anal. Calcd. for C18H,00*N6SCl: C, 43.5; H, 8.1. Found: C, 43.0; H, 8.3.
We gratefully acknowledge a grant from Sharp and Dohme which made possible a part of this work.
summary Some new derivatives of thiamorpholine-3,5dicarboxylic acid have been prepared, several of them having interesting pharmacological potentialities. NEWHAVEN,CONN.
ROCKEFELLER INSTITUTE
FOR
RECEIVED APRIL 24, 1940
MEDICAL RESEARCH]
The Structure of Trimethyl Glucurone BY RICHARD E. REEVES' Pryde and Williams2 have prepared from glucurone a crystalline trimethyl derivative which they regard to be 1,2,4-trimethylglucurone(I). Although no evidence substantiating this structure has been presented, Marrack and Carpenter3 also refer to the trimethylglucurone as having this structure. Such a substance on hydrolysis and oxidation might be expected to yield 2,4-dimethylsaccharic acid, which is wanted as a reference compound in this Laboratory. When the trimethylglucurone, m. p. 129-130°, was prepared, its properties were found t o resemble those of a methylfuranoside more than the methylpyranoside called for by structural formula I. On acid hydrolysis reducing sugar was liberated a t the rapid rate characteristic of methylf ~ r a n o s i d e s . ~Pryde and Williams have noted (1) Present address: Boyce Thompson Institute for Plant Research, Inc., Yonkers, N. Y. (2) J. Pryde and R. T. Williams, Biockem. J., 17, 1203 (1933). (3) J. Marrack and B. R. Carpenter, Brit. J. Exptl. Path., 19, 59 (1938). (4) W. N. Haworth and E. L. Hirst, J. Chem. SOL,2616 (1930).
the rapid fall in rotation when trimethylglucurone is dissolved in methanol containing hydrogen chloride. It is now shown that this fall in rotation is accompanied by the formation of an isomeric trimethylglucurone, m. p. 90-91') having a low rotation. This suggests the rapid mutarotation which Levene and Meyer5 have observed for CYand 0-methylfuranosides under similar conditions. The high melting trimethylglucurone was hydrolyzed and oxidized to a dimethylsaccharic acid which was characterized as the crystalline diamide. When the crude dimethylsaccharic acid was esterified with cold ethereal diazomethane, no crystalline dimethylsaccharic acid esters or lactone esters were obtained; instead there was produced an unsaturated dimethyl lactone methyl ester (11) known to have its methyl groups on positions 2 and 5.6 This (5) P. A. Levene and G.M. Meyer, J . B i d . Chcm., 74,701 (1927). (6) 0.T h . Schmidt, H. Zeiser and H. Dippold, Bev., 70, 2402 (1937).
STRUCTURE OF TRIMETHYL GLUCURONE
June, 1940
finding, together with the above-mentioned properties, make it appear that the furanoside structure 111, 2,5-dimethyl-a-methylglycosideof glucurone, is probably the correct structure for the trimethylglucurone of Pryde and Williams. The low melting isomer and the diamide are regarded as 2,5-dimethyl-~-methylglycoside of glucurone and 2,5-dimethylsaccharic acid diamide, respectively.
1617
by the modified Hagedorn-Jensen p r o c e d ~ r e . ~The results are given in cc. of 0.01 N thiosulfate required by 1 cc. of hydrolysate after refluxing for a stated length of time: Omin.,O.llcc.; 5min.,1,76cc.; 15min.,3.91cc.; 30min., 4.65 cc.; 60 min., 4.49 cc. To minimize the effect of the gradual decomposition of reducing sugar a curve was drawn through the points and the velocity constant was calculated by the method of Guggenheims for the first twenty minutes of the hydrolysis. The value was found to be k = 1380 X 10-6 m h - 1 for N/lOO HCI at loo", which is definitely within the range exhibited by most furanosides. When recalculated oa the basis of natural logarithms the values found by Haworth and Hirst4 for furanosides in 0.01 N acid a t 100" range from 600 X to 11,000 x 10-6. 2,5-Dimethylsaccharic Acid Diamide.-Trimethylglucurone (100 mg.), m. p. 129-130', was heated for four hours a t 80-85' in 4 cc. of dilute nitric acid, sp. gr. 1.2. Nitric acid was removed by evaporating in vacuo, adding water and repeatedly evaporating to dryness in vacuo. The residue was then esterified by heating for three hours a t 75" in a sealed tube with 4 cc. of methanol containing 2% dry hydrogen chloride, The hydrogen chloride was removed with silver carbonate. Attempts to obtain crystals from this ester both before and after vacuum distillation were unsuccessful. The sirup was dissolved in 2 cc. of methanol, cooled to O", and treated with ammonia gas. Crystals separated, and more were obtained by the addition of ether to the mother liquors; 43 mg. of crude material yielded 11mg. of pure diamide, m. p. 169-170°, after three recrystallizations from ethanol. Anal. Calcd. for CsH1606N2: C, 40.68; H, 6.78; N, 11.88; CH30, 26.3. Found: C, 40.62; H, 6.77; N, 11.70; CH30, 26.46. Unsaturated Saccharolactone Methyl Ester 11.-Two hundred mg. of high-melting trimethylglucurone was hydrolyzed and oxidized with nitric acid as in the preparation of the diamide. The saccharic acid after removal of nitric acid and water was treated with 15 cc. of an 0.5 molar solution of diazomethane in ether, and allowed to stand a t 0" for three days. The solution was filtered from a small flocculent precipitate, and concentrated to 3 cc. Forty mg. of crystals separated, which on recrystallization from acetone-ether melted at 85-86.5". Mixed with the unsaturated lactone ester, m. p. 87-88", prepared from saccharolactone methyl ester9 the melting point was 8687 '.
Hf=7 7 ; FTH f r'" 1 CHOCH,
COOCHi
HCOCHs I
HCOCHa
0
-CH
HCOCHs
?
H
A
I
' -co / I
L
H CI
1
\OCH,
1
HSOCHa H C A
-co
L-co I1
Experimental Part
I I
HCOCH3
I11
High-Melting Trimethylg1ucurone.-Pryde and Williams' report a 25y0yield of this substance following three riiethylations of glucurone with silver oxide and methyl iodide. By the present procedure a 50% yield was obtained after one methylation. To 5.11 g. of glucurone was added 15 cc. of acetone and 20 cc. of methyl iodide. The vigorously stirred mixture was cooled in an ice-bath during t h e addition of 3 g. of silver oxide. The temperature was gradually raised to 45" with the addition of 27 g. of silver oxide during nine hours. Fifteen cc. of methyl iodide and 10 cc. of acetone were added and the mixture was then refluxed at 4 5 4 8 " for five hours. The product was extracted with chloroform and crystallized from acetone by the addition of ether, a second and third crop being obtained from the mother liquors. After recrystallization 2.71 g. of 2,5-dimethyl-cr-methylglycosideof glucurone, m. p. 129-130°, was obtained; sp. rot. (D-line) (24') 151' (c, 0.4 in CHClJ. Low-melting Trimethylg1ucurone.-Trimethylglucurone, m. p. 129-130°, (372.6 mg.) was dissolved a t 23' in 15 cc. of methanol containing 3.6% dry hydrogen chloride. After seventy minutes the rotation (1-dm. tube) became constant a t -0.03". The solution was evaporated to dryness in vacuo and, when the residue was redissolved in summary methanol and again taken to dryness, crystals formed in the distillation flask. Recrystallization from methanol Evidence is presented which indicates that the and ether gave 270 mg. of 2,5-dimethyl-~-methylglycoside trimethylglucurone prepared by Pryde and Wilof glucurone, m. p. 90-1", sp. rot. (D-line) (21") 2.0" (c, liams is the 2,5-dimethyl-a-methylglycosideof 1.0 in HzO); (24') -2.3" (c, 0.9 in CHCls). glucurone. Anal. Calcd. for C~H1406: C, 49.53; H, 6.42; CH,O, 42.6. Found: C, 49.24; H, 7.15; CHaO, 42.85. NEWYORK,N. Y. RECEIVED MARCH29,1940 Kinetic8 of Hydrolysis.-Trimethylglucurone (94.6 mg.) (7) C. S. Hanes, Biocltcnz. J . , 43, 99 (1929). m. p. 129-130" was refluxed in 50 cc. of 0.05 N hydrochloric (8) E. A. Guggenheim. Phil. Mag., [7] 9, 538 (1926). acid and the reducing sugars were determined a t intervals (9) R. E. Reeves, THISJOURNAL, 61, 664 (1939).
